Zhouyang Liu's homepage

About Me

I am a Postdoctoral Researcher at the National University of Defense Technology (NUDT). I received my Ph.D. in Computer Science from NUDT under the supervision of Prof. Dongsheng Li (李东升).

I work on graph learning, with current interests in open-world graph learning, Graph + LLM, and AI for Science. Before that, I spent quite some time on graph similarity computation.

I am always happy to chat about research ideas — polished or half-baked. If you are interested in collaborating, just drop me an email.

Prospective students: If you are interested in working with me, send me your CV and a few lines about what you would like to work on.

Research

My research interests include:

Open-World Graph Learning
Graph + LLM
Graph Similarity
AI for Science
AI Security

Recent Updates

[Oct. 2026] We released GSCBench, a benchmark for evaluating the generalization of neural graph edit distance models across graph collections. [Paper] [Code] [Data]
[Sep. 2026] FlexST, our work on multi-domain pre-training for universal traffic forecasting, was accepted to IEEE ICDM 2026. [Paper] [Code]
[May 2026] GEN, our work on one-step alignment for flexible graph similarity computation, was published at IEEE ICDE 2026. [Paper] [Code]
[Jan. 2026] I joined NUDT as a Postdoctoral Researcher.